Promoting physical activity and decreasing sedentary behavior are key goals in the fight against cancers; physical activity is associated with lower risk of several cancers [1-10], and lower overall morbidity and mortality [11-26]. Thus, theory-driven initiatives to change these behaviors are essential [1-10, 26-40]. PQ#3 highlights the necessity for new perspectives on the interplay of cognitive and emotional factors in promoting behavior change. Current theories, which focus primarily on predictors derived from self-report measures, do not fully predict behavior change. For example, recent meta-analyses suggest that on average, variables from the Theory of Planned Behavior account for ~27% of the variance in behavior change [41, 42]. This limits our ability to design optimally effective interventions [43], and invites new methods that may explain additional variance. Our team has shown that neural activation in response to health messages in hypothesized neural regions of interest can double the explained variance in behavior change, above and beyond self-reports of attitudes, intentions, and self-efficacy [44, 45]. We now propose a next leap, inspired by PQ3, to identify how cognitive and affective processes interact in the brain to influence and predict behavior change. Our core hypothesis is that the balance of neural activity in regions associated with self-related processing versus defensive counterarguing is key in producing health behavior change, and that self-affirmation (an innovative approach, relatively new to the health behavior area [46]) can alter this balance. Self-affirmation theory [47] posits that people are motivated to maintain a sense of self-worth, and that threats to self-worth will be met with resistance, often i the form of counterarguing. One common threat to self-worth occurs when people are confronted with self-relevant health messages (e.g. encouraging less sedentary behavior in overweight, sedentary adults). This phenomenon speaks to a classic and problematic paradox: those at highest risk are likely to be most defensive and least open to altering cancer risk behaviors [48]. A substantial, and surprisingly impressive, body of evidence demonstrates that affirmation of core-values (self-affirmation priming) preceding messages can reduce resistance and increase intervention effectiveness [46, 49-53]. Uncovering neural mechanisms of such affirmation effects [46], has transformative potential for intervention design and selection. To test our conceptual assumptions and core hypothesis we will: (1) Identify neural signals associated with processing health messages as self-relevant versus counterarguing; (2) Test whether self-affirmation alters the balance of these signals; (3) Use these neural signals to predict physical activity behavior change, above and beyond what is predicted by self-report measures alone. Our approach is innovative methodologically (using fMRI to understand and predict behavior change), and conceptually (self-affirmation may dramatically increase intervention effectiveness). Benchmarks will include objectively measured decreases in sedentary behavior in affirmed vs. control subjects (using accelerometers), and increases in predictive capacity afforded by neuroimaging methods, compared to self-report alone.